ParCorr: efficient parallel methods to identify similar time series pairs across sliding windows - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Journal Articles Data Mining and Knowledge Discovery Year : 2018

ParCorr: efficient parallel methods to identify similar time series pairs across sliding windows

Djamel-Edine Edine Yagoubi
  • Function : Author
  • PersonId : 1086292
Reza Akbarinia
Boyan Kolev
Oleksandra Levchenko
Florent Masseglia
Patrick Valduriez

Abstract

Consider the problem of finding the highly correlated pairs of time series over a time window and then sliding that window to find the highly correlated pairs over successive co-temporous windows such that each successive window starts only a little time after the previous window. Doing this efficiently and in parallel could help in applications such as sensor fusion, financial trading, or communications network monitoring, to name a few. We have developed a parallel incremental random vector/sketching approach to this problem (as explained in section 4) and compared it with the state-of-the-art nearest neighbor method iSAX [7]. Whereas iSAX achieves 100% recall and precision for Euclidean distance, the sketching approach is, empirically, at least 10 times faster and achieves 95% recall and 100% precision on real and simulated data. For many applications this speedup is worth the minor reduction in recall. Our method scales up to 100 million time series and scales linearly in its expensive steps (but quadratic in the less expensive ones).
Fichier principal
Vignette du fichier
ParCorr__DMKD2018.pdf (637.24 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

lirmm-01886794 , version 1 (03-10-2018)

Identifiers

Cite

Djamel-Edine Edine Yagoubi, Reza Akbarinia, Boyan Kolev, Oleksandra Levchenko, Florent Masseglia, et al.. ParCorr: efficient parallel methods to identify similar time series pairs across sliding windows. Data Mining and Knowledge Discovery, 2018, 32 (5), pp.1481-1507. ⟨10.1007/s10618-018-0580-z⟩. ⟨lirmm-01886794⟩
265 View
438 Download

Altmetric

Share

Gmail Facebook X LinkedIn More